全内反射椭偏成像（TIRIE）生物传感器是将光学椭偏成像技术工作在全内反射条件下，并结合微流道反应器阵列形成的生物传感器，可以实现生物分子相互作用的实时检测，具有无标记、实时检测、高通量、灵敏度高等优点，已成功用于生物医学、临床诊断以及环境监测等领域。但其对于氨基酸、核苷酸等生物小分子的低浓度检测还存在灵敏度不足的问题，而这些小分子的超灵敏检测对于生物医学、药物学、环境安全等方面具有重要的意义。

为实现TIRIE传感器对于生物小分子的超灵敏检测，需提升传感系统的灵敏度和信噪比。因此，本论文以提升TIRIE传感系统的灵敏度为研究目标，对超灵敏TIRIE生物传感方案进行研究。本文主要从优化光学元件的偏振设置和降低系统噪声两个方面来提升传感器的信噪比。

首先，从理论上探究了系统的光学响应信号与光学元件的偏振设置之间的关系，并在消光¾非消光条件以及非消光条件下优化了光学元件的偏振设置，给出了两种条件下光学响应信号的变化与椭偏参数的变化之间的关系。另一方面，分析了现有传感系统的噪声源，指出噪声主要来源于光源、检测池和探测器。并针对不同的噪声源，提出了一些抑制噪声的技术和方法：采用参比通道以及空间平均的方法来降低光源功率噪声；采用控温装置以及数值补偿的方法来降低温度波动噪声；通过对探测器进行制冷来降低探测器噪声；以及在传感界面处施加交流电势，并通过频谱分析的方法滤波降噪。最后，利用TIRIE传感器对分子量为174Da的精氨酸进行了检测，试验结果表明，在优化光学元件的偏振设置及降低系统噪声后，该传感系统可以实现生物小分子的超灵敏检测。

Total internal reflection imaging ellipsometry (TIRIE) biosensor is formed by combining imaging ellipsometry operated in the total internal reflection mode and a multichannel micro-fluidic reactor array. TIRIE biosensor can monitor the real-time biomolecule interactions, and has the advantages of label-free, real-time detection, high throughput and high sensitivity. And it has been successfully used in biomedicine, clinical diagnosis and environmental monitoring. However, its sensitivity is not enough to detect small biological molecules, such as amino acids and nucleotides. The detection of these small biological molecules is of great significance in biomedicine, pharmacology and environmental safety.

In order to realize the detection of small biological molecules by TIRIE biosensor, it is necessary to improve the sensitivity and signal-to-noise ratio (SNR) of TIRIE system. Therefore, this paper aims to improve the sensitivity of TIRIE system, and studies the scheme of ultra-sensitive TIRIE biosensor. This paper mainly improves the SNR of TIRIE system by optimizing the polarization settings of the polarizer and the analyzer and reducing the system noises.

Firstly, the relationship between the optical responses of the system and the polarization settings of the optical elements is investigated theoretically. The polarization settings of optical elements are optimized under the null-off null working condition and the off null working condition. And the relationship between the variation of optical responses and the variation of ellipsometric parameters under the two conditions is analyzed. On the other hand, the noise sources of TIRIE system are analyzed. The system noises mainly come from the light source, the detection cell and the photoelectric detector. According to the different noise sources, some noise suppression techniques are proposed. The power noise of light source is reduced by using the reference channel and spatial averaging method. The temperature fluctuation noise is reduced by using the thermostatic apparatus and numerical compensation method. The detector noise is reduced by cooling the CCD detector. The periodic potential disturbance is applied on the sensing surface, and the noise is filtered by spectrum analysis. Finally, arginine with a molecular weight of 174 is tried to be detected by using TIRIE system. The experimental results show that after optimizing the polarization settings of the optical elements and reducing the system noises, the TIRIE system can realize the ultra-sensitive detection of small biological molecules.